Model comparison
Same task, different trade-offs.
We are not looking for an absolute winner. The comparison shows which model better suits your work, budget and required speed.
Models being compared
GPT-6 Luna
Data from September 27, 2026
Published September 22, 2026
A cost-effective option for clearly defined, repetitive tasks.
Gemini 3.8 Flash
Data from September 27, 2026
Published September 2, 2026
Processes text, images, audio and video; the output is text.
Price, speed and usage
$0.10 input / $0.50 output per million tokens. Above 272 thousand input tokens: 2× input and 1.5× output.
$0.75 input / $3.75 output per million tokens through December 31, 2026. From January 1, 2027: $1.50 / $7.50.
We have not independently measured speed on a comparable basis.
We have not independently measured speed on a comparable basis.
OpenAI API
Gemini API and Google AI Studio
1,050,000 tokens in total; output capped at 128 thousand tokens.
1,048,576 input tokens; output capped at 65,536 tokens.
Text and images → text
Text, images, video, audio and PDF → text
GPT-6 Luna
Use when…
- Text classification and brief summaries
- Processing a large number of simple requests
Choose another model when…
- For complex tasks, check whether the low price leads to more corrections.
Gemini 3.8 Flash
Use when…
- Analysis of documents, audio and video
- Programming and multi-step business tasks
Choose another model when…
- A different model is needed to generate images or audio directly.
Original sources